Go Back  DVD Talk Forum > General Discussions > Tech Talk
Reload this Page >

Some thoughts on the state of the web (kinda long, hopefully interesting)

Tech Talk Discuss PC Hardware, Software, Internet and Other Technology

Some thoughts on the state of the web (kinda long, hopefully interesting)

Old 09-02-05, 08:42 AM
  #1  
DVD Talk Legend
Thread Starter
 
Join Date: Oct 1999
Location: |-|@><0r [email protected]|)
Posts: 17,214
Some thoughts on the state of the web (kinda long, hopefully interesting)

I picked up O'Reilly's Programming ASP.NET recently. It's now collecting dust next to the other, let's see, one two three four books on ASP.NET on my O'Reilly Zoo bookshelf.

ASP.NET is a technology that I want to love. The concept of embedding controls on a web page, just as you would a Windows application, carries web programming into the 21st century - it gets away from crappy, declarative programming in HTML via that state-of-the-art development tool called Notepad. It's an idea whose time has come. Because in a nutshell, a user interface is a user interface - shouldn't matter whether it's embedded in an OS window or in a browser. Same textboxes, same buttons, same lists and drop-downs, etc. It should all be the same shit. Web programming should only differ where it has to differ - like making a trip back to the server for a new page, or some data exchange.

So I have a soft spot for the theory of ASP.NET. The problem is that it's still crappy. The "ASP server controls" library is weak - instead of mirroring the Windows application control set, they cobbled together some junk that's a touch better than an HTML button control, but not by much. The ASP controls look different; they have far fewer attributes; you can't anchor them. It's stupid. And the developer still spends too much time on extraneous web-programming crap, like embedding some script elements into the HTML by hand.

But that's not the worst. You want to know what's the worst? Here's what's worst.

One of the superb features of C# is the ease of drawing within a control. You can just plop down a picture box, and hook it up to an OnPaint function that scribbles all over it with pens, brushes, polygons, bitmaps, whatever. And it's easy either to hook up the redraw to a periodic timer - "tell this control to paint itself every 50 milliseconds," maybe - or to force a redraw in another part of the code, like in a button-click handler. It's easy, and the result and performance are both good (neither of which could be claimed about the predecessor API, the Windows GDI.) All of the neural network applications I wrote for the MCIS program make extensive use of these painting functions.

So you'd think they would port this capability into ASP.NET - that they would extend actual program functionality to the web. It wouldn't be hard - just build some capability into the browser to run IL code in a fast, tight virtual machine. Security is an issue - you don't want code embedded in a web page messing with your CD-ROM drive, of course - but the .NET people already thought through this issue, in excruciating detail, with the code access security. But you'd be wrong.

Notice that advances in Internet technologies, including Microsoft's, have kind of stagnated over the last two or three years? What's new recently - VoIP, and, uh, that's about it, right? Why is this? It's not because the technology has reached its potential - anyone with a sense of things believes that the Internet is in its infancy - or because funding has dried up. It's because the technology behind the web has not matured. Websites remain, in essence, HyperCard on steroids. Well, that's not quite fair; HyperCard was not only a neat concept, but was a way-easy development platform. In fact, it's gotten much, much worse. Ten years ago, we had HTTP, HTML, Gopher, telnet, FTP, mail, IRC, and Perl acting as a CGI gateway. Maybe you'd do some funky VRML vectoring, if you wanted to be all cutting-edge. Today, on top of those old and stale techologies, we have layered: DHTML, CSS, Java, Flash, PHP, JavaScript, JScript, VBScript, ASP and the quite different ASP.NET, FrontPage extensions, XML, web services, Python, ActiveDirectory, network domain and security issues, VPN, SSL, content management systems, and MySql and SQL Server back-ends. These really aren't new capabilities or new features; they're new ways of doing the same shit. The web has become a hideous swamp of technologies blighted with interoperability issues and version incompatibilities. (If you're thinking of a career in web development, don't. The pain and headaches are not worth it.)

Microsoft is well-positioned to solve this problem. The effective organization in .NET could be extended to web technologies. So why is Microsoft not doing this?

I think the answer is apparent. If programs run in web browsers as well as on Windows, why buy Windows? Just run IE on your Linux or whatever. It's just thin-client architecture. Due to Microsoft's worship of its golden goose (the desktop OS monopoly), it has developed a crippling paranoia of any threat, including (especially) the age of the thin, non-Windows client. And so, web development continues to suck, and gets worse with every Hot New Technology.

Here's the problem with this: Microsoft really doesn't have this option.

Someday, someone will get web programming right. We will look back on Java, Flash, and other attempts as crude precursors, the FORTRANs and COBOLs of web programming. Someone will find a way to turn web browsers into full-featured computer interfaces. The front runner is Google, and they're amply sharp enough to succeed. But someone will.

Microsoft can either be that guy, or cede the OS market to that guy. It doesn't have the option of squashing the web while continuously scraping the rust off of Windows x+1 and repackaging it.

It would be amazing for Microsoft to miss this boat. I recently read the insightful point that Microsoft began as a programming-languages company: its first product was a BASIC interpreter for an Altair 8080. Programming should be its bread and butter. The OS thing is really just a shell; and to any techies who've read the press releases about Windows Vista, it looks like Microsoft has milked that cow to death. There will be no large, impressive advances in operating systems for the next many years. Where Microsoft can innovate is on the web. And it's not doing that.

Well, that's the way the tech industry goes. The future will be interesting.

- David Stein
sfsdfd is offline  
Old 09-12-05, 07:51 PM
  #2  
DVD Talk Limited Edition
 
Join Date: Jun 2000
Location: BV VA
Posts: 6,061
What they could be, should be, but will be doing is whatever makes the most money at the time. Unfortunately the masses like to see new operating systems because it's different and new. As long as it looks newer and still functions as good as the last released you'll have buyers. Sucks but thats big business
kantonburg is offline  
Old 09-12-05, 08:01 PM
  #3  
X
Administrator
 
X's Avatar
 
Join Date: Oct 1987
Location: AA-
Posts: 10,704
You can do what you want with ActiveX control programming. Of course that's only going to run with IE. The .NET framework actually makes it fairly efficient.

But in general ASP.NET controls suck in terms of functionality due to the stateless form of web pages.
X is offline  

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is On
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Thread Tools
Search this Thread

Archive Advertising Cookie Policy Privacy Statement Terms of Service

Copyright 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.